Home Affront: Housing Across the Generations

REPORT WEALTH SERIES

September 2017 Adam Corlett & Lindsay Judge HOME AFFRONT Housing across the generations

intergencommission.org @resfoundation Home Affront Acknowledgements 2

Acknowledgements

The authors would like to thank the many experts who have shared their thoughts on this report through conversations and comments including Kate Barker, Jonathan Halket, Neal Hudson, John Muellbauer, Joe Oldman, Bob Pannell, Kate Webb, Sally West and Peter Williams. The UK Data Archive also helped with historical data. Any errors or omissions are, of course, our own.

This work contains statistical data from ONS which is Crown Copyright. The use of the ONS statistical data in this work does not imply the endorsement of the ONS in relation to the interpretation or analysis of the statistical data. This work uses research datasets which may not exactly reproduce National Statistics aggregates.

intergencommission.org @resfoundation Home Affront Contents 3

Contents

Acknowledgements �� 2

Executive summary �� 4

Section 1: Introduction ��8

Section 2: Housing security over time ��9

Section 3: Housing affordability over time ��29

Section 4: Housing quality over time ��43

Section 5: Looking to the future ��53

Section 6: Conclusion ��62

Annex 1: Our housing tenure categories and data ��63

Annex 2: Measuring affordability ��66

Annex 3: Modelling of homeownership rates ��67

intergencommission.org @resfoundation Home Affront Executive summary 4

Executive summary

Many disparities exist between the generations but none is more acutely felt than the question of housing. Across all income groups and all parts of the country, young people believe they are on the receiving end of a poorer housing experience than their parents and grandparents. Moreover, this angst transcends the generations: both older family members and the broader population flag housing as one of their key concerns for young people in the UK today.

In this, the 9th report of our Intergenerational Commission, we take on this hugely important topic. We compare the housing outcomes achieved by different generations over the life course and assess the extent to which intergenerational inequalities exist when it comes to security, to affordability and to quality. We explore how the housing experience of each generation has been determined by demographics, policy and the market alike. Finally we look to the future, recognising that our housing tenure by mid-life can have a significant bearing on our living standards not just in the here and now but also over the entire span of our lives. So what do we find when we take a long, hard look at housing across the generations?

Today’s thirty year olds are only half as likely to own their home as the baby boomers

Britain may traditionally be viewed as a nation of home owners but a look at the data shows that the share of families who own their home has been falling since 2003. This downward trend has not been felt equally across the generations: home ownership rates of older birth cohorts have remained steady at high levels while younger generations have seen much lower levels of, and slower increases in, ownership than their predecessors. As a result, today’s families headed by 30 year olds are only half as likely to own their home as their parents were at the same age.

Demographic changes such as longer education, later coupling and later child bearing form part of our account as to why younger people’s home ownership rates have fallen so sharply over time. Moreover, we note that the home ownership rates of older generations received a boost in the 1980s and 1990s as a result of Right to Buy (and the decision not to replenish the social housing stock means fewer properties available for young people to rent in this sector today).

Taken together, however, preferences and policies like Right to Buy provide only part of the explanation as to why home ownership rates have declined over time. Crucially, we show how barriers to entry have increased dramatically, not least from rising house prices. With the average young family today having to save for 19 years to accumulate enough for a typical deposit compared to just 3 years a generation ago, it is small wonder that home ownership rates have tumbled.

intergencommission.org @resfoundation Home Affront Executive summary 5

Young people today are four times as likely to rent privately than they were two generations ago

Falling home ownership and a shrinking social rented sector have together conspired to increase the ranks of private renters over time. As well as functioning as the residual tenure for those who cannot or do not want to live in these two other main types, we show that policy decisions about rents, notice periods and lending have also driven the growth of the sector. As a result, four out of ten 30 year olds live in private rented accommodation today in contrast to one in ten 50 years back, giving rise to the term ‘generation rent’.

While usually an acceptable tenure when footloose and fancy free, private renting can become far less desirable over the life course if it brings with it considerable insecurity. We show that the low security of tenure which is a hallmark of private renting today - and which critically makes it so much less attractive to families as they age and have children - is a historical development. Tenants have much stronger rights in many other countries and indeed had more security in the UK before 1988, albeit not without some associated problems.

Private renting is not just the least secure form of housing, however, but also the tenure with the worst record for housing quality. We show that private renters have historically lived in poorer conditions than those in other tenure types, and that it is the oldest tenants today who are most likely to live in non-decent homes. But with so many young people renting privately today, the sheer size of this group living in the worst quality tenure should give policy makers pause for thought.

Families often house their adult children – but increasingly not their parents

For all the talk of the ‘boomerang-ers’ we see that adult children living with parents is a common phenomenon over time. That said, the proportion of children who live with their parents beyond their majority does fluctuate, in part in response to economic conditions (we saw a rise in rates when unemployment increased at the start of the 1980s, for example) and in part as behaviours change (the increased number of young adults in higher education has had a clear effect on rates). Whether this way of living is a choice or a necessity, we note that close to half of today’s 25 year olds who are single and without children currently reside in the parental home.

From an intergenerational point of view a far more striking development is that families are much less likely to house their parents than they have in the past. While this no doubt speaks to welcome improvements in longevity and better health into older age, we might also speculate that changing social norms – likely including the dramatic rise in female employment rates over the long term - provide another part of the explanation.

Generation on generation, housing costs have absorbed a larger share of family income with significant living standards effects

While much of the focus is on declines in home ownership, the direct effect of housing on living standards is much bigger than simply what tenure people live in. The average share of income that families spend on housing has trebled over the last 50 years. The

intergencommission.org @resfoundation Home Affront Executive summary 6

last half a century has seen rising housing costs increasingly acting as a drag on living standards, leaving many – including the old but especially the young – facing an affordability crisis today. On average millennials spend 23 per cent of their income on housing compared to the 17 per cent baby boomers spent at the same age, and the 8 per cent of the silent generation.

Growing affordability differentials between the tenure types is a key driver of these trends. Affordability has always varied between tenures, but the gap has increased significantly. Throughout the 1960s and 1970s, mortgaged owners spent around 5 per cent of their income on housing costs, renters writ large around 10 per cent. Today, that wedge has grown: while mortgaged owners spent around 12 per cent of their income on their housing costs in 2016, private renters were paying three times as much (36 per cent) with massive implications for the intergenerational experience.

There are some winners when looking at housing affordability across the generations however. Today’s millennials who by hook or by crook have purchased a home early in their lives have lower interest costs when measured as a share of income than mortgaged owners in the previous two generations at the same age. This is largely driven by record low interest rates alongside the fact that barriers to entry increasingly make home ownership a better-off young person’s game. But the proportion of income being spent on capital repayments has risen relentlessly from generation to generation. This, coupled with the possibility that the near-zero interest rate environment will not endure – should temper our assessment of home owning millennials as the lucky few.

Housing stock quality has improved, but the young are making compromises on space and commuting

Housing costs may have risen as a share of income over time but isn’t this simply because we get more for our money? In many respects this may be true: mass slums, outside toilets and inadequate running water are thankfully things of the past (for most). But while rising housing affordability has occurred alongside significant improvements in housing stock for older generations over the course of their lives, today’s young people are being squeezed literally as well as financially. Millennial-headed households are more likely than previous generations to live in overcrowded conditions for example, and when we look at the distribution of square meterage we see today’s under-45s have been net losers in the space stakes compared to previous cohorts, while over-45s are net gainers.

Younger people today appear to be compromising not just on housing quality but also on quality of life. When we look at travel to work times we note that millennials have longer commutes than older generations did at the same age. What may look like only small increases in the travel on a daily basis add up over time: if the current differences we observe between the average commuting times of each generation were to endure, we estimate that millennials will spend 64 hours (close to three full days) more commuting in the year they turn 40 than their parents did at the same age.

Even in a best-case scenario millennials will not achieve the same home ownership levels the baby boomers enjoy

Throughout this report we note that tenure is often (although not always) the critical determinant of the housing outcomes we can achieve. While there is no ‘right’ level of home ownership, we recognise that this remains the preferred tenure of most in the UK, and has manifold advantages beyond the simple provision of housing.

intergencommission.org @resfoundation Home Affront Executive summary 7

Given this, we end this report with an examination of the ownership prospects of young people today. While we expect home ownership to pick up in the next few years as we move away from the stresses of the financial crisis, there is significant uncertainty about the scale of any recovery. Drawing on data from 1961 onwards we construct a model that explains how prices, incomes, credit and supply have conspired at different points in time to produce different home ownership levels, and use this to produce upper and lower estimates of the future home ownership rates of millennials.

We find that if the underlying conditions that prevailed in the decade with the strongest home ownership growth (1981-1991) were replicated, the share of young people set to ultimately own their homes would reach similar levels to generation X by the age of 45 but remain around 6 percentage points lower than that of the baby boomers (our ‘optimistic’ scenario). There are downsides even to this rosy of account however: deferred home ownership will increase the likelihood of living in rented accommodation when raising children for example, while paying off a mortgage later in life could constrain other forms of saving.

Conversely, if the experience of the poorest performing home ownership decade (2002-2012) were to be repeated we estimate that less than half of millennials will buy a home before the age of 45 compared to over 70 per cent of baby boomers who had done so by that age (our ‘pessimistic’ picture). Such a scenario could have wide ranging consequences politically (with renting as the majority tenure, could radical reform of the private rented sector be on the cards?) and financially (increasing the amount spent on housing over the entire life course as well as hindering wealth accumulation).

That all said, an open question remains as to what could happen to younger cohorts’ home ownership rates as older generations age and die. Will this wealth be consumed during later life or absorbed by social care costs? Or could it be bequeathed and thereby resolve the housing issue for some families (albeit perhaps at a point in the life course when the beneficiaries are no longer raising children)? In future analysis for the Inter- generational Commission we will look at the question of the timing and distribution of inheritance in detail.

Housing is a majoritarian concern – but it is young people who are at the sharp end of the housing crisis

For better or for worse, the big housing trends of the last half a century – tenure change, rising costs, rising quality – affect us all, but the impact of each has fallen on the different generations in very different ways. For many older people affordability increases earlier in their lives went hand in hand with improved security (especially rising home ownership) as well as vast improvements in the housing stock. In contrast, many of today’s young people are getting less for their money whether we look at their housing experience in terms of space, security or quality of life.

We should not be without hope, however. Throughout this examination of the intergenerational housing experience we have seen time and again that policy matters. Just what history – and other countries – can teach us about what we could do to address the housing catastrophe of the last fifty years will be a topic we will return to once again in the final months of our Intergenerational Commission.

intergencommission.org @resfoundation Home Affront Section 1 8

Section 1

Introduction

In our most recent Resolution Foundation report for the Intergenerational Commission we saw that a staggering half of people today expect younger generations to fare less well in the future than their predecessors.1 And across the board - old and young, high or low income - the number one concern that informed this sense of pessimism was housing.

In this, the ninth report for our Intergenerational Commission, we turn our attention to this critical topic. While we have analysed housing trends over time in previous work, for the first time here we present our analysis using an intergenerational lens.2 We explore how long term housing trends such as tenure change, rising costs and improved quality have affected different birth cohorts over the life course, resulting in quite different outcomes when it comes to security, living standards and quality of life over the generations. To this effect:

• Section 2 explores how each generation has fared in terms of housing security by looking at the prevalence of different tenures as well as the degree of security within each over time.

• Section 3 then turns to the question of affordability and considers how ongoing housing costs have borne down on incomes over the past 50 years.

• Section 4 analyses the changes we have witnessed in housing quality over half a century, showing how housing stock has improved considerably over this time but that there are other ways that housing can impact on the quality of life.

• Section 5 then looks to the future, exploring the factors that have determined home ownership rates historically and using this data to model two possible home ownership scenarios for young people today.

• Section 6 offers someconcluding thoughts.

• And for those who would like more details of our findings, the Annexes provide information on methodology and technical specifications.

1 H Shrimpton, G Skinner and S Hall, The millennial bug: Public attitudes on the living standards of different generations, Resolution Foundation 2017 2 S Clarke, A Corlett and L Judge, The housing headwind: The impact of rising housing costs on UK living standards, Resolution Foundation 2016

intergencommission.org @resfoundation Home Affront Section 2 9

Section 2

Housing security over time

Studies have long shown that housing can have different meanings over place and time beyond the simple provision of shelter. It can be a way of signalling status, for example, or a key method of storing and accumulating wealth. For most people, however, housing primarily means a home. Crucially it is the sense of ‘security’ we experience – how much freedom we have to live freely in a property alongside how much control we have about when to move on – that determines whether or not we regard it as such. In this section we explore the level of housing security that different generations have enjoyed at points of their life course and show how policy and demographics have both played key roles in explaining the differential outcomes we observe.

We do this largely through an analysis of tenure – the legal rights we exercise in relation to the property we call home.3 As Table 1 shows, tenure may not be a perfect proxy for housing security – there are clearly sources of security and insecurity attached to every tenure type. Moreover, and particularly relevant for our intergenerational analysis, the level of security attached to what is nominally the same tenure (for example, private renting) can change over time.

Table 1: Security and tenure Securities Insecurities Built up an asset which can be borrowed against, Outright ownership Responsible for maintenance and repairs drawn down on or bequeathed Full enjoyment of property

Building up an asset which can be borrowed Mortgaged ownership At risk of repossession if cannot repay mortgage against, drawn down on or bequeathed Full enjoyment of property subject to mortgage Responsible for maintenance and repairs terms (e.g. may not be able to sub-let)

Right to quiet enjoyment (e.g. landlord cannot Can be asked to leave without reason with two Private renting (assured shorthold tenancy) enter property without notice) months notice Often restrictions on use e.g. no pets, not able to Landlord should keep property in good repair change decoration

Cannot be evicted unless landlord applies for Private renting (regulated or controlled Right to quiet enjoyment (e.g. landlord cannot possession order from courts which can only be tenancy) enter property without notice) granted in certain cirumstances Often restrictions on use e.g. no pets, not able to Landlord should keep property in good repair change decoration

Right to quiet enjoyment (e.g. landlord cannot Social tenancies increasingly not offered on life Social renting enter property without notice) time terms Landlord should keep property in good repair Limited rights of succession

Living with family or friends At family's discretion At family's discretion

Such provisos apart, however, tenure trends provide a good starting point for our ex- ploration of housing security – allowing us to explore how the generations have fared over many decades rather than just a single point in time.

3 See Annex 1 for technical details on how we have defined tenures throughout this section

intergencommission.org @resfoundation Home Affront Section 2 10

Tenure trends writ large

We begin our investigation with Figure 1, which shows the distribution of families in the UK by tenure since 1961 (see Box 1 for further discussion about why we chose ‘the family’ as our preferred unit of analysis throughout this report).

Figure 1: Proportion of families (singles or couples) by tenure over time: UK Technical chart info (esp y axis) 100% Full-time student in parents' home 90% Single adult in parents' home 80% In someone else's owned or social rented home 70% Sharing private rent 60% Private rent alone 50% Housing association rent 40% Council rent 30% Own with mortgage 20% Own outright 10% Change of data source 0% 1961 1966 1971 1976 1981 1986 1991 1996 2001 2006 2011 2016

Note: Data source switches in 1984. Full-time students are not separated from other single adults living in their parents’ home prior to 1984. Housing associations are included within the private rented sector prior to 1980. Sample size is limited in 1961-1967. See Annex 1 for more details

Source: RF analysis of Family Expenditure Survey 1961-1983; Labour Force Survey 1984-2017

As Figure 1 makes clear, on this aggregate level there are three key tenure stories that emerge from analysis of the last half a century (and one non-story).

• The first story relates to home ownership. The proportion of families owning their home grew steadily up to 2003, reaching an all-time high of 58.3 per cent.4 However, while the proportion owning their home outright (i.e. without a mortgage) continues to break records today, the proportion buying with a mortgage has been falling. Indeed, this downward trend started in 1996, with only the larger increase in outright ownership keeping overall home ownership on an upward path up to of 2003. Also worth noting is the fact that the downward trend in overall home ownership has recently halted, with rates edging up by around 0.6 percentage points over the past year.

4 See estimates for home ownership in 1918, for example, in Office of National Statistics, ‘Home ownership down and renting up for first time in a century’, ONS June 2015 intergencommission.org @resfoundation Home Affront Section 2 11

• The second trend we can pick out relates to the rise and fall of the social rented sector over the last 50 years. The proportion of families housed in this tenure reached a peak of 29 per cent in 1981, but today this stands at a record low (14 per cent). The mix within the sector has also shifted. Housing associations have substituted for some, but certainly not all, local authority tenancies and now account for a large part of the provision in this sector.

• The third pattern we note relates to the fall and rise of the private rented sector. The proportion of families in this tenure shrank from 28 per cent in the early 1960s to just 8 per cent in 1989, before rising again to stand at 18 per cent in 2016. The proportion of renters who share with other families has fluctuated over time, but this pattern is more pronounced for private renters who live in single benefit unit households.

• The non-story we can tell relates to the proportion of adults living in a family member’s home. At this aggregate level this figure is lower than it was at the beginning of our time series although it has varied over the period.

Figure 2: Proportion of families age 25-34 by tenure over time: UK

Technical100% chart info (esp y axis) Single adult in parents' home 90% In someone else's owned or 80% social rented home Sharing private rent 70% Private rent alone 60%

50% Housing association rent

40% Council rent

30% Own with mortgage

20% Own outright 10% Change of data source 0% 1961 1966 1971 1976 1981 1986 1991 1996 2001 2006 2011 2016

Note: Data source switches in 1984. Housing associations are included within the private rented sector prior to 1980. Sample size is limited in 1961-1967. Here, we have not separated out full time students from other adult singles living at home as the former are relatively few in number for this age group. See Annex 1 for more details.

Source: RF analysis of Family Expenditure Survey 1961-1983; Labour Force Survey 1984-2017

intergencommission.org @resfoundation Home Affront Section 2 12

When we look at how tenure has changed for younger families over time, our three tenure stories stand in even starker relief and our non-story changes somewhat. As Figure 2 shows, the fall in home ownership, the decline of the social rented sector, the expansion of private renting (and especially sharing within it), and the rise in the proportion of adults living in their parents’ homes in recent years are all trends that are heavily inflected by age.

i Box 1: Measuring tenure

Those who follow such things may note that the the three renting residents drop out of the data. Or proportions we show in Figure 1 look quite different from consider an adult returning to the parental home or an the conventional rates of home ownership and renting that older person moving in with younger family. If we use the are frequently cited in academic studies and the press. household as our unit of analysis those individuals simply disappear from the statistics. Likewise, if five unrelated To date, tenure has largely been measured by assessing people shared a house they would be counted as one how many dwellings fall into each tenure type. But this rented household rather than the five separate renters potentially provides a misleading picture as it only tells us that most would intuitively regard them to be. the proportion of properties which are owned or rented. While that may have some macro-economic significance, All these real-life situations have largely been missed out for those who are interested in living standards – as well of the tenure story to date. As a result, throughout this as suppressed demand for housing – it makes far more report we have switched our attention from dwellings or sense to think instead about the proportion of families households to families (by which we mean the benefit who live in each tenure type. unit comprising a single adult or a couple, along with any dependent children they may have living with them). Imagine, for example, a person who buys a house but who Following this approach, Figure 1 therefore provides a then takes in three lodgers. On the standard measure, much more nuanced and accurate picture of how we have this simply counts as an owner-occupied household – and been housed over time.

intergencommission.org @resfoundation Home Affront Section 2 13

The baby boomers have benefited most from the rise in home ownership

In Figure 3 we follow the approach taken in other Intergeneration Commission work to look at the home ownership rates of different generations and then compare how each has fared at the same age.5 As the chart makes clear, the baby boomer generation has enjoyed a higher home ownership rate from early adulthood right through to retirement than any generation before or since. In contrast, millennials’ home ownership rates lag not just their parents and their grandparents, but also their great-grandparents at the same age. In fact, today’s young people are around half as likely to own their home at the age of 30 as their baby boomer counterparts at the same age.6

Figure 3: Generational home ownership rates by age: UK Technical chart info (esp y axis) 80%

70%

60%

50%

40%

30%

20% Millennials (1981-2000) Gen X (1966-1980) Baby boomers (1946-1965) Silent gen (1926-1945) 10% Greatest gen (1911-1925) Forgotten gen (1896-1910) Lost gen (1881-1895) 0% 20 25 30 35 40 45 50 55 60 65 70 75

Note: Figures for each generation are derived from a weighted average of estimates by single year of age for each single-year birth cohort within that generation; generations are included if at least five birth years are present in the data

Source: RF analysis of Family Expenditure Survey 1961-1983; Labour Force Survey 1984-2017

5 See, for example, L Gardiner, Stagnation generation: The case for renewing the intergenerational contract, Resolu- tion Foundation 2016 6 Note that for each generation not all members will have reached every age. So for example, we will not be able to definitely say what the Generation X home ownership rate at 45 is until 2025, when those born in 1980 have reached that age

intergencommission.org @resfoundation Home Affront Section 2 14

Figure 4: Five-year cohort home ownership rates by age: UK Technical chart info (esp y axis) 80%

70% 1946-1950 60% 1951-1955

50% 1956-1960

1961-1965 40% 1966-1970 30% 1971-1975

20% 1976-1980

1981-1985 10% 1986-1990 0% 20 25 30 35 40 45 50 55 60 65 70

Note: Results are shown only where every member of a cohort has reached that age.

Source: RF analysis of Family Expenditure Survey 1961-1983; Labour Force Survey 1984-2017

Figure 4 presents a more finely grained version of this story by looking at the home ownership rates of five-year birth cohorts from the baby boomers onwards (with results shown only where every member of a cohort reached that age). As this makes clear, it is the oldest boomer cohorts who have been the real beneficiaries of the growth in home ownership over the last 50 years with each subsequent cohort trailing its predecessor.

intergencommission.org @resfoundation Home Affront Section 2 15

… While young people writ large have seen home ownership rates fall

Declining home ownership is not a minority problem for today’s young people. In Figure 5 we focus our attention on the oldest millennial cohort (born 1981-85) and look at how home ownership has changed by region when we compare them to an earlier cohort (born 1961-1965). While their rates vary dramatically across regions (from more than half owning in the North West to less than a quarter doing so in Inner London for example), home ownership has fallen in every single region (with the fall in youth home ownership in Outer London particularly notable).

Figure 5: Home ownership rates by region: age 25-34, selected regions only Technical chart info (esp y axis) 60%

50%

40%

30%

20% Strathclyde Rest of North West

Wales South East 10% Tyne & Wear West Midlands (met county)

Outer London Inner London 0% 1984 1986 1988 1990 1992 1994 1996 1998 2000 2002 2004 2006 2008 2010 2012 2014 2016

Source: RF analysis of Labour Force Survey 1984-2017

We can repeat the exercise for the same cohorts but this time splitting our results by income quintiles. Figure 6 presents the picture and shows once again that the falls in home ownership have been felt across the distribution with rates roughly halving for all groups since mid-1990s.

intergencommission.org @resfoundation Home Affront Section 2 16

Figure 6: Home ownership rates by household income quintile, age 25-34: GB Technical chart info (esp y axis) 80%

70%

60%

50%

40%

30%

20% Richest quintile 2nd richest quintile 10% Middle quintile 2nd poorest quintile Poorest quintile 0% 1961 1965 1969 1973 1977 1981 1985 1989 1993 1997 2001 2005 2009 2013

Note: 1994-2015 data based on HBAI and not consistent with the LFS data used elsewhere in this report.

Source: RF analysis of Family Expenditure Survey 1961-1991; Households Below Average Income 1994-2015

Changing demographics explain part – but far from all – of the millennials’ home ownership lag

How fair is it, however, to compare the home ownership rates of a 30 year-old baby boomer with a 30 year-old today? There is an extensive literature that shows how housing choices are intimately related to life course events, with the transition from renting to home ownership particularly strongly correlated with partnering and having children.7 Moreover, extended education and especially the significant increase in the number of young people today who are educated to degree level means that by the age of 30 many millennials have spent significantly less time in the labour market than their grandparents had by a similar age (albeit commanding a higher wage when they do start working).

What would happen to home ownership rates if we adjusted the composition of today’s young people to match the same levels of labour market exposure, partnering and child

7 See for example, D Clapham, The meaning of housing: A pathways approach, University of Bristol 2005. See also, however, C Askoy, Short term effects of house prices on birth rates¸ EBRD working paper 2016 which suggests a potential circularity with its finding that higher housing costs may be contributing to the rising age at which women have their first child

intergencommission.org @resfoundation Home Affront Section 2 17

bearing that we observed in the same age groups in 1984/1985? Figure 7 provides the results for this simple thought experiment, charting the actual observed home ownership rates for our various age cohorts in 1984/1985 and 2016, and the home ownership rates for each in 2016 reweighted on the assumption that their demographic profile was the same as that of the equivalent age band in 1984/1985.8

Figure 7: Home ownership rates by age band, 1984-1985 and 2016 (actual and reweighted) Technical chart info (esp y axis) 2016 actual 2016 with 1984/1985's demographics 1984/1985 actual

24-26 years

27-29 years

30-32 years

33-35 years

36-39 years

40-42 years

0% 10% 20% 30% 40% 50% 60% 70%

Source: RF analysis of Labour Force Survey

For those aged 30-32, these selected demographic differences do indeed explain some of the gap we see between today’s home ownership rates and those observed in 1984/1985: roughly one-third in total. However, these compositional effects appear to diminish as we move up the age scale, all but wearing off by the age of 40. What’s clear is that there is more to generation-on-generation declines in home ownership at any given age that simple shifts in demographics.

8 To do this we follow the approach developed in J Browne, Reweight2: Stata module to reweight survey data to user-defined control totals, IFS, July 2012

intergencommission.org @resfoundation Home Affront Section 2 18

Older generations have benefited from social housing twice over

Home ownership is not the only way that it has been possible to enjoy a high level of security of tenure over the last 50 years. The post-war social housing project was explicitly designed to offer families a place they could rent, but nonetheless call their own. While lifetime tenancies were only officially introduced with theHousing Act 1980, prior to this the sufficiency of stock meant that a socially rented property could effectively be regarded as one’s permanent home. As Figure 8 shows, older generations were considerably more likely to live in social housing than younger people today.

Figure 8: Generational social renter rates by age: UK Technical chart info (esp y axis)

35%

30%

25%

20%

15%

10% Millennials (1981-2000) Gen X (1966-1980) Baby boomers (1946-1965) Silent gen (1926-1945) 5% Greatest gen (1911-1925) Forgotten gen (1896-1910) Lost gen (1881-1895) 0% 20 25 30 35 40 45 50 55 60 65 70 75

Note: Prior to 1980, housing associations are included in the private rented sector. Also see Figure 3

Source: RF analysis of Family Expenditure Survey 1961-1983; Labour Force Survey 1984-2017

The Housing Act 1980 introduced another way that older generations may have benefited from social renting – namely Right to Buy. The selling off of social housing at discounts of between a third and half on market prices both reduced social renting rates (see Figure 9) while boosting the home ownership rates of the generations before the baby boomers. While Right to Buy remains in place (except in Scotland from August 2016) and could be extended to housing association tenants in the near future, a range of factors (for

intergencommission.org @resfoundation Home Affront Section 2 19

example, diminished stock of social housing, higher house prices, lower discounts and a greater concentration of low income families in social housing today) suggest that this is not a policy from which younger generations will equally benefit.

Figure 9: Generational social renter rates by year: UK

Technical45% chart info (esp y axis) Millennials (1981-2000) Housing Act 1980 Gen X (1966-1980) 40% Baby boomers (1946-1965) Silent gen (1926-1945) Greatest gen (1911-1925) 35% Forgotten gen (1896-1910) Lost gen (1881-1895) 30%

25%

20%

15%

10%

0% 1961 1966 1971 1976 1981 1986 1991 1996 2001 2006 2011 2016

Note: See Figure 3

Source: RF analysis of Family Expenditure Survey 1961-1983; Labour Force Survey 1984-2017

A final point worth noting in relation to the generational differences faced by social housing tenants is that secure tenancies which can effectively last a lifetime are increasingly becoming a thing of the past. The Localism Act 2011 gave local authorities and housing associations the power to make new letting on a less secure footing, with DCLG data showing that 16 per cent of new tenancies in England were on a non-lifetime basis in 2015-16.9 However, the limited take up of this discretionary power led to further changes being made in the Housing and Planning Act 2016, with social landlords in the future required to offer fixed term tenancies except in very exceptional circumstances.

9 Department for Communities and Local Government, Social housing lettings, England April 2015 to March 2016, DCLG 2016

intergencommission.org @resfoundation Home Affront Section 2 20

The inexorable rise of private renting

Locked out of home ownership and less able to access social housing than previous generations, today’s young people are unsurprisingly far more likely to live in the private rented sector than in the past. Figure 10 shows the scale of this change: at the age of 30 only one in ten baby boomer families rent privately, compared to two in ten of the silent generation and generation X. Today millennials are four times as likely to be privately renting at this age, with all the insecurities attached to this tenure type.

Figure 10: Generational private renter rates (all) by age: UK

Technical chart info (esp y axis) Millennials (1981-2000) Gen X (1966-1980) 40% Baby boomers (1946-1965) Silent gen (1926-1945) Greatest gen (1911-1925) Forgotten gen (1896-1910) 35% Lost gen (1881-1895)

30%

25%

20%

15%

10%

0% 20 25 30 35 40 45 50 55 60 65 70 75

Note: Prior to 1980, housing associations are included in the private rented sector. Also see Figure 3

Source: RF analysis of Family Expenditure Survey 1961-1983; Labour Force Survey 1984-2017

So what explains the trends we see in private renting over time? In many accounts the size of the sector is very much treated as the residual – it is the tenure for those who cannot or do not wish to own or rent socially. However, when we look across the generations as we do in Figure 11 we can also discern the effect of public policy choices on its evolution. Rent controls and strong tenants’ rights in the post-war period made it less attractive to set up as a private landlord, while the introduction of mortgage interest relief at source (MIRAS) for home owners in 1983 also tilted the balance in favour of home ownership during the earlier part of our time series. But the phasing out of rent control after the Housing Act 1988, the introduction of Buy-to-Let mortgages in 1996

intergencommission.org @resfoundation Home Affront Section 2 21

(which accelerated particularly steeply after 2001)10 and the abolition of MIRAS from 2000 have all clearly played a role in reinvigorating the private rented sector with some dramatic intergenerational effects.11

Figure 11: Generational private renter rates by year: UK Technical chart info (esp y axis) 45% Millennials (1981-2000) Housing Act Gen X (1966-1980) 1988 40% Baby boomers (1946-1965) Silent gen (1926-1945) BTL 1996 35% Greatest gen (1911-1925) Forgotten gen (1896-1910) End of MIRAS Lost gen (1881-1895) 2000 30%

25%

20%

15%

10%

0% 1961 1966 1971 1976 1981 1986 1991 1996 2001 2006 2011 2016

Note: See Figure 3

Source: RF analysis of Family Expenditure Survey 1961-1983; Labour Force Survey 1984-2017

Moreover, as Table 1 intimated renting privately today does not come with the same protections as it did three decades ago. As Figure 12 shows, the majority of tenures were either regulated or controlled in the 1960s through to the 1980s. As a result, landlords would struggle to evict tenants unless there was a breach of contract, and tenancies could even be passed on after death to partners living in the property. The introduction

10 National Housing and Planning Advice Unit, Buy-to-let mortgage lending and the impact on UK house prices: a technical report, February 2008 11 See also L Gardiner, ‘Home sweet homes: The rise of multiple property ownership in Britain’, Resolution Foun- dation August 2017 which shows that rental properties are mainly owned by baby boomers and generation X

intergencommission.org @resfoundation Home Affront Section 2 22

of the assured shorthold tenancy in 1988 marked a sea change for renters however: those with such contracts (the majority since at least 1994) could be given just two months’ notice to vacate.

Figure 12: Types of private tenancies: England Technical chart info (esp y axis) 4.5m Assured shorthold Assured 4.0m Other Not accessible to public 3.5m Regulated, rent not registered Regulated, rent registered 3.0m Controlled 2.5m

2.0m

1.5m

1.0m

0.5m

0.0m 1964 1967 1970 1973 1976 1979 1982 1985 1988 1991 1994 1997 2000 2003 2006 2009 2012

Notes: Figure for 1964 includes Wales. Reweighting in 2001-02 produces a small discontinuity. 2014-15 total (matching EHS Table 1.1) is produced on a different basis to earlier years.

Source: DCLG Survey of English Housing (Table S510), DCLG English Housing Survey Private Rented Sector Report, 2014-15 and Holmans’ Historical Statistics of Housing in Britain (Table G.12)

Younger private renters are more likely to share – in part (but not entirely) because of migration

There have been some striking changes over time when we focus our attention on those who share in the private rented sector rather than rent as a single benefit unit. Figure 13 presents the headlines for this sub-group. Here, unlike in the overall private rent picture (Figure 10) we observe sharp reductions in sharing across the life course for the most recent generations. However, when we look back to the greatest generation and before, we find relatively high levels of shared renting prevailing even in middle and later age.

intergencommission.org @resfoundation Home Affront Section 2 23

Figure 13: Generational shared private renter rates by age: UK Technical chart info (esp y axis) 25% Millennials (1981-2000) Gen X (1966-1980)

Baby boomers (1946-1965) Silent gen (1926-1945) 20% Greatest gen (1911-1925) Forgotten gen (1896-1910)

Lost gen (1881-1895) 15%

10%

0% 20 25 30 35 40 45 50 55 60 65 70 75

Note: Prior to 1980, housing associations are included in the private rented sector. See Figure 3

Source: RF analysis of Family Expenditure Survey 1961-1983; Labour Force Survey 1984-2017

There are two interesting things to note when we scratch beneath the surface. First, in the 1960s sharing in the private rented sector was very much the preserve of older people; today the age profile of sharers has completely changed. Second, millennials renting ‘alone’ (i.e. as a single benefit unit) are far more likely to be in couples than in preceding generations.

It is also important to note both the compositional and indirect effects that immigration has had on the private rented sector and housing tenure more broadly, particularly given that this has been more important for some generations (with those born outside the UK accounting for one in four millennial families now in the 30-35 range). Increases in net migration from around 1997, and particularly with the A8 accession in 2004, meant increased demand for housing and contributed to a rise in the private rented sector – and shared renting especially – as Figure 14 shows.

intergencommission.org @resfoundation Home Affront Section 2 24

Figure 14: Numbers of family units headed by someone born outside the UK by tenure type: UK Technical chart info (esp y axis) 5.0m Single adult in parents' home In someone else's owned or social rented home 4.0m Sharing private rent

Private rent alone 3.0m Housing association rent

2.0m Council rent

Own with mortgage

1.0m Own outright

Discontinuity in numbers 0.0m 1984 1987 1990 1993 1996 1999 2002 2005 2008 2011 2014 2017

Note: Discontinuity in absolute numbers between 2014Q4 and 2015Q1

Source: RF analysis of Labour Force Survey

Boomeranging back with a vengeance?

At the start of this Section we could see that it has not been unusual historically for adult children to live with their parents (Figure 1). So why all the fuss about young people remaining at home today?

Figure 15 looks at the age profile of each cohort of adult children who reside with parents, noting that this is primarily – though not exclusively – an under-30s phenomena. At around ages 20-21, millennials have been less likely to live in their parents’ home than the preceding cohorts, most likely because of increased numbers living in private accommodation while at university (but not those living in halls, whom we have excluded from our analysis). However, at ages 23-29, millennials have been more likely than generation X and baby boomers to be living at home – lending some credence to the concept of a so-called ‘boomerang generation’.

intergencommission.org @resfoundation Home Affront Section 2 25

Figure 15: Five-year cohort rates of adults living in parental home: UK Technical chart info (esp y axis) 90% 1946-1950 1951-1955 80% 1956-1960 1961-1965 70% 1966-1970 1971-1975 60% 1976-1980 1981-1985 50% 1986-1990 1991-1995 40%

30%

20%

10%

0% 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35

Note: Excludes young people who themselves have children or live as a couple in the parental home and full-time students.

Source: RF analysis of Family Expenditure Survey 1961-1983; Labour Force Survey 1984-2017

So what could account for this change? Figure 16 shows the profile of adults living in the parental home and from this we can make three key observations. First, the rising number of adults in full time education provides part of the explanation as to why we have seen figures rise in recent years. Second, we see some degree of association with economic conditions – the number of adults living at home rose when unemployment increased at the start of the 1980s and again when incomes fell around the financial crisis. And third, those living at home (excluding full-time students) are disproportion- ately male (with men accounting for about two-thirds of the total).

intergencommission.org @resfoundation Home Affront Section 2 26

Figure 16: Number of single childless adults (19+) living in their parents’ home: UK Technical chart info (esp y axis) 4,500,000

4,000,000 All (inc. FT students) 3,500,000 All (ex. FT students) 3,000,000

2,500,000 Men (ex. FT students) 2,000,000

1,500,000 Women 1,000,000 (ex. FT students)

500,000 FT students (male & female) 0 1961 1965 1969 1973 1977 1981 1985 1989 1993 1997 2001 2005 2009 2013 2017

Note: Students in halls of residence are not included in this data

Source: RF analysis of Family Expenditure Survey 1961-1983; Labour Force Survey 1984-2017

Today’s older people are less likely to live with family than they did in the past

Finally in this Section, it is worth considering the inverse experience: when parents come back to live in their children’s homes. Here we see another stark change between the generations with Figure 17 illustrating how it has become increasingly rare for older relatives to live with their younger family.12 In part of course this speaks to welcome improvements in longevity and better health into older age, but we might also speculate that changing social norms provide another part of the explanation.

12 Note that those living with family in the private rented sector are not captured here but instead appear as private renting sharers

intergencommission.org @resfoundation Home Affront Section 2 27

Figure 17: Generational rates of living in someone else’s owned or social rented home (rolling average over three years of age): UK

Technical16% chart info (esp y axis) Millennials (1981-2000) Gen X (1966-1980) Baby boomers (1946-1965) Silent gen (1926-1945) 14% Greatest gen (1911-1925) Forgotten gen (1896-1910) Lost gen (1881-1895) 12%

10%

0% 20 25 30 35 40 45 50 55 60 65 70 75 80

Note: See Figure 3. Generations are included if at least three birth years are present in the data

Source: RF analysis of Family Expenditure Survey 1961-1983; Labour Force Survey 1984-2017

The other side of this coin is presented in Figure 18. It shows the proportions of owners and social renters who have shared their home with someone other than their single adult children across the generations. As we saw above, baby boomers have been more likely in their 50s and 60s to share their home with their children than previous generations were. But this chart shows that the opposite is true in relation to other adults, with a trend for each generation in middle-age to be less likely to share their home with other family.

Although the need for many people to care for older relatives is often discussed – with much talk of a ‘sandwich generation’ of families simultaneously having children and older relatives living with them – this move away from sharing a home is noteworthy. The shift has no doubt also been influenced by – and/or has influenced – the increase in female employment from the silent generation to the baby boomers to generation X.13

13 See, for example, L Gardiner, Stagnation generation: The case for renewing the intergenerational contract, Resolution Foundation 2016

intergencommission.org @resfoundation Home Affront Section 2 28

Figure 18: Generational rates of sharing owned or social rented home with someone other than single adult children (rolling average over three years of age): UK

Technical chart info (esp y axis) 9%

Millennials (1981-2000) Gen X (1966-1980) 2% Baby boomers (1946-1965) Silent gen (1926-1945) Greatest gen (1911-1925) Forgotten gen (1896-1910) 1% Lost gen (1881-1895)

0% 20 25 30 35 40 45 50 55 60 65 70 75

Note: See Figure 3. Generations are included if at least three birth years are present in the data

Source: RF analysis of Family Expenditure Survey 1961-1983; Labour Force Survey 1984-2017

Taken together, tenure trends mean than today’s young people enjoy substantially less housing security than previous generations

All in all, the multiple changes in tenure that we have outlined in the Section add up to a toxic combination for young people when compared them to their parents and grandparents. Millennials are far less likely to own their own home at the age of 30 than previous generations; they are less likely to be social renters today or have a secure tenancy if they are; and they are far more likely to rent privately but have less legal protection than the renters of yesteryear. Demographic changes and associated shifts in preferences may inform some of the tenure trends we observe, and policy has a strong bearing on the outcomes that generations can achieve at specific points in time. But we have yet to consider how much it costs to achieve the security that most regard as the defining feature of a home. It is to the question of housing affordability that we now turn.

intergencommission.org @resfoundation Home Affront Section 3 29

Section 3

Housing affordability over time

In the previous section we saw how demographic changes and policy decisions have shaped the security that families have been able to enjoy in their homes over time. Next we turn to the topic of housing affordability. We consider both how housing costs have weighed on the living standards of different generations over their life course, and the implications of affordability for the housing options that are over time are open to specific groups.

Measuring housing affordability is not straightforward however. While attention is often focused on headline house prices or rents, it is the relationship between a family’s ongoing housing costs and their disposable income that, in our view, best captures housing affordability.14 As a result, in this section and elsewhere, we use the housing cost to income ratio (HCIR) as our primary affordability metric. For example, if a couple has an after-tax income of £20,000 and spends £6,000 a year on housing, their HCIR would be 30 per cent.15 So what do we find when we look at housing affordability over time? Housing cost to income ratios have trebled over half a century Figure 19 presents the historical picture for all families alongside those living in each of the main tenure types. As this makes plain, the overall trend is one of significant increases in housing costs relative to income. While the average family spent just 6 per cent of their income on housing costs in 1961, today this has trebled to 18 per cent.16

14 See S Clarke, A Corlett & L Judge, The housing headwind: The impact of rising housing costs on UK living standards, Resolution Foundation, June 2016 for a further discussion of measuring housing affordability 15 See Annex 2 for full details of methodology we use in this section 16 While we do not currently have data for 1992 and 1993 there is no reason to think there is a discontinuity between the 1991 and 1994-95 data. See, for example, C Bellfield, D Chandler and R Joyce, Housing: Trends in Prices, Costs and Tenure, IFS 2015

intergencommission.org @resfoundation Home Affront Section 3 30

Figure 19: Average housing cost to income ratio by tenure type: GB

Technical40% chart info (esp y axis) Private renters 35% Housing association renters 30% Local authority renters 25% All 20% Mortgagors (exc. principal) 15% Outright owners 10%

0% 1961 1965 1969 1973 1977 1981 1985 1989 1993 1997 2001 2005 2009 2013

Note: Income and housing costs both include housing benefit. 2016-17 nowcast. 1992 and 1993 are interpolated. See Annex 2 for further details

Source: RF analysis of IFS HBAI (FES) 1961 to 1991; DWP HBAI (FRS) 1994-95 to 2015-16

Increases have been recorded across most tenures, but there has been a striking divergence in the scale of these between those who own and those who rent. Among private renters for instance, the average HCIR has increased from 9 per cent at the start of the 1960s to 36 per cent today. In contrast, the average HCIR among outright owners is just 5 per cent. That’s still a doubling from the figure of 2.5 per cent in 1961, but it is just a fraction of the ratio recorded among renters. Moreover, it is worth noting the relative decline in the average HCIR after 1990 among those buying a property with a mortgage – with a particularly marked reduction from 2009. So what has driven these startling trends? To begin, the HCIRs of mortgaged families are largely determined by the interplay between house prices at the point of purchase and interest rates over time. For renters, we need to look not just to the market but also to policy for an explanation. Deregulation of private rents from the 1980s has clearly driven up the HCIR of tenants in the private rented sector over time, and also had a feedthrough effect on social rent levels which are set in part with reference to prevailing market conditions.17 Moreover, the proportion of income that social renter families spend on housing has risen not just as a result of higher social rents (the numerator of the HCIR), but also because their average income (the denominator) has fallen since 1980 as many better-off tenants moved into home ownership via Right to Buy.

17 See, for example, W Wilson, Rent setting: Social housing (England), House of Commons Library 2017

intergencommission.org @resfoundation Home Affront Section 3 31

Before concluding that this pattern of rising HCIRs across all tenures is unequivocally a ‘bad’ thing for living standards, we should note that housing is a ‘superior’ good. As such, on one argument we should naturally spend a greater proportion of our income on housing costs over time as our incomes rise.18 Relatedly, we might be spending more on housing simply because we are getting a better product today than we used to, a question we consider in some detail in Section 4. For the remainder of this section however, our starting point is that a rising HCIR represents a headwind to living standards.

Today’s 30 year olds spend significantly more on housing than their grandparents did at the same age

Figure 20 moves beyond the headline HCIR data to give us to get a better sense of how housing affordability has varied across the life course for different generations, with two things standing out. First, there is a marked difference between the shape of the life course curves experienced by the older and younger generations: the silent generation was the first to have a curve that peaks during working-age rather than continuing to rise into retirement. The second trend of note relates to the fact that the proportion of income allocated to housing costs has increased generation-after-generation over the entirety of the period.

Figure 20: Proportion of income spent on housing costs by generations: GB

Technical25% chart info (esp y axis)

20%

15%

10%

5% Millennials (1981-2000) Gen X (1966-1980) Baby boomers (1946-1965) Silent gen (1926-1945) Greatest gen (1911-1925) Forgotten gen (1896-1910) Lost gen (1881-1895) 0% 20 25 30 35 40 45 50 55 60 65 70 75

Note: Income and housing costs both include housing benefit. Incomes and housing costs are assumed to be shared equally within households. Figures for each generation are derived from a weighted average of estimates by single year of age for each single-year birth cohort within that generation; generations are included if at least five birth years are present in the data.

Source: RF analysis of IFS HBAI (FES) 1961 to 1991; DWP HBAI (FRS) 1994-95 to 2015-16

18 See, for example, A Turner, Between debt and the devil: Money, credit and fixing global finance, Princeton 2016 intergencommission.org @resfoundation Home Affront Section 3 32

Interestingly, the jump experienced between the silent generation and the baby boomers was particularly marked for those aged roughly 25-45. In contrast, the step change occurring between baby boomers and generation X was more marked at younger ages. With average HCIRs rising again among millennials, we are left with a picture in which today’s 30 year olds spend a significantly larger share of their income on housing than their grandparents and great-grandparents did at the same age. Figure 21 provides more detail, by looking at the HCIRs of five year birth cohorts. The step changes in HCIRs during working-age years remain visible, though they are less marked over the most recent four cohorts. Much clearer in this chart, however, is the extent to which ratios appear to converge across cohorts in later life. It remains to be seen if such convergence continues as younger cohorts reach retirement. Even if it does though, the cumulative effect of the significantly steeper curves recorded over the working-age part of younger groups’ life courses will have significant implications for overall life time housing costs.

Figure 21: Proportion of income spent on housing costs by five year birth cohorts: GB Technical chart info (esp y axis) 25%

20%

15%

10%

5% 1926-1930 1931-1935 1936-1940 1941-1945 1946-1950 1951-1955 1956-1960 1961-1965 1966-1970 1971-1975 1976-1980 1981-1985 1986-1990 0% 20 25 30 35 40 45 50 55 60 65 70

Note: See Figure 20

Source: RF analysis of IFS HBAI (FES) 1961 to 1991; DWP HBAI (FRS) 1994-95 to 2015-16

intergencommission.org @resfoundation Home Affront Section 3 33

As ever, these aggregates cover a multitude of experience. We have already seen the differing HCIR trends recorded across tenures at the headline level; what then do our generational patterns look like when we look through a tenure-specific lens?

Housing Benefit mitigates the affordability differences for social renters across the generations Turning first to social renters, we observe in Figure 22 a subtly different pattern to the overall one depicted in Figure 20. We still see each generation reporting higher HCIRs than the generation before them when compared at the same ages (social renting baby boomers had a HCIR of 15 per cent at the age of 30 for example, compared to 27 per cent for millennials and gen-Xers). What’s different, however, is that social renters of all ages today are spending a similar amount of their income on rent (around 30 per cent).

Figure 22: Proportion of income spent on housing costs by generations, social renters only: GB

Technical35% chart info (esp y axis)

30%

25%

20%

15%

10%

Note: See Figure 20

Source: RF analysis of IFS HBAI (FES) 1961 to 1991; DWP HBAI (FRS) 1994-95 to 2015-16

intergencommission.org @resfoundation Home Affront Section 3 34

In fact, the remaining intergenerational differences diminish even further when we look at social renters’ HCIRs net of Housing Benefit (i.e. deducting it from both the income and housing costs) as we do in Figure 23. While there is still evidence of some generation-on-generation increase in HCIRs, it is much more muted. Missing from this picture however is any data covering the period before the 1980s, when housing subsidy was mainly provided in the form of low rents rather than Housing Benefit.19

Figure 23: Proportion of income on housing costs by generations net of housing benefit, social renters only: GB

Technical18% chart info (esp y axis)

16%

14%

12%

10% Millennials (1981-2000) Gen X (1966-1980) 8% Baby boomers (1946-1965) Silent gen (1926-1945) 6%

0% 20 25 30 35 40 45 50 55 60 65 70 75

Note: All specifications same as Figure 20 except income and housing costs are net of HB. Generational curves here contain data from 1994 onwards only

Source: RF analysis of DWP HBAI (FRS) 1994-95 to 2015-16

Members of Generation Rent spend twice as much of their income on housing costs as the baby boomers did in their youth The picture looks somewhat different for private renters, with generational patterns again differing subtly from the trends set out in Figure 2. The first thing to notice in Figure 24 is that private rents take up a much greater proportion of income for young people today than they did for previous generations: baby boomer renters spent one-fifth of their income on housing costs at the age of 30 compared to one-third spent by 30 year olds today. The especially big jump at younger ages between baby boomers and generation

19 See K Webb, Bricks to Benefits: Rebalancing housing investment, Shelter 2012 for a full account of this shift

intergencommission.org @resfoundation Home Affront Section 3 35

X is likely to largely flow from the ending of ‘fair rents’ policies that dominated in the 1970s and earlier. Coupled with the fact that so many more young people live in the private rented sector today than did in the past (see Section 2), it is no surprise that we increasingly talk about millennials as ‘Generation Rent’.

Figure 24: Proportion of income on housing costs by generations, private renters only: GB

Technical50% chart info (esp y axis) Millennials (1981-2000) Gen X (1966-1980) Baby boomers (1946-1965) Silent gen (1926-1945) 45% Greatest gen (1911-1925) Forgotten gen (1896-1910) Lost gen (1881-1895) 40%

35%

30%

25%

20%

15%

10%

0% 20 25 30 35 40 45 50 55 60 65 70 75

Note: See Figure 20

Source: RF analysis of IFS HBAI (FES) 1961 to 1991; DWP HBAI (FRS) 1994-95 to 2015-16

Once again, the picture is altered once we account for Housing Benefit. Figure 25 shows that private renters’ HCIRs are reduced across the board when we measure incomes net of Housing Benefit, but in contrast to social renters a clear generation-on-generation increase remains in place.

intergencommission.org @resfoundation Home Affront Section 3 36

Figure 25: Proportion of income on housing costs by generations net of housing benefit, private renters only: GB

Technical40% chart info (esp y axis)

35%

30%

25%

20%

15% Millennials (1981-2000) Gen X (1966-1980) 10% Baby boomers (1946-1965) Silent gen (1926-1945) 5%

0% 20 25 30 35 40 45 50 55 60 65 70 75

Note: All specifications same as Figure 20 except income and housing costs are net of HB. Generational curves here contain data from 1994 onwards only

Source: RF analysis of DWP HBAI (FRS) 1994-95 to 2015-16

The lucky few: millennial owners

In stark contrast to the experience of renters, cost to income ratios among millennial home owners with mortgages have fallen relative to those recorded at the same age by baby boomers and members of generation X. As Figure 26 shows, the reduction at age 30 is substantial: whereas baby boomer and generation X owners spent close to 20 per cent of their income on housing costs, the equivalent figure among millennials is 15 per cent.

intergencommission.org @resfoundation Home Affront Section 3 37

Figure 26: Proportion of income on housing costs by generations, mortgaged owners only: GB

Technical20% chart info (esp y axis)

18%

16%

14%

12%

10%

4% Millennials (1981-2000) Gen X (1966-1980) Baby boomers (1946-1965) Silent gen (1926-1945) 2% Greatest gen (1911-1925) Forgotten gen (1896-1910) Lost gen (1881-1895) 0% 20 25 30 35 40 45 50 55 60 65 70 75

Note: See Figure 20

Source: RF analysis of IFS HBAI (FES) 1961 to 1991; DWP HBAI (FRS) 1994-95 to 2015-16

There are of course several reasons why the picture may not be quite as rosy as it first seems. To begin, millennial HCIRs remain much higher than those recorded among owners in the silent generation: the big step change that occurred between the silent generation and the baby boomers has been only partially unwound. Secondly, we know that home ownership rates have fallen among millennials, meaning that the group enjoying this apparent improvement relative to generation X represents a minority. Thirdly, the housing market that most millennial owners have entered has been characterised by unusually low interest rates (with this backdrop also reflected in sharp falls in HCIRs among older generation X and baby boomer owners). This last point raises an important question about what this chart will look in the future if – or when – interest rates rise.

Delving beyond the generational averages allows us to see the interest rate effect more clearly. Figure 27 shows the same results but for five-year birth cohorts and indicates that the groups that faced the toughest mortgage interest affordability in early adulthood were the 1956-60 and 1961-65 cohorts. Someone born in 1963 would have been 27 in 1990 when the base interest rate hit 14.875 per cent. (It is no coincidence that mortgage arrears and repossessions skyrocketed in the early 1990s recession). In contrast, today’s young people benefit from record low interest rates with the bank rate currently at 0.25 per cent.

intergencommission.org @resfoundation Home Affront Section 3 38

Figure 27: Proportion of income spent on housing costs by five year birth cohorts, mortgaged owners only: GB Technical chart info (esp y axis)

25%

20%

15%

10%

5% 1926-1930 1931-1935 1936-1940 1941-1945 1946-1950 1951-1955 1956-1960 1961-1965 1966-1970 1971-1975 1976-1980 1981-1985 1986-1990 0% 20 25 30 35 40 45 50 55 60 65 70

Note: See Figure 20

Source: RF analysis of IFS HBAI (FES) 1961 to 1991; DWP HBAI (FRS) 1994-95 to 2015-16

That said, young mortgagors average HCIRs are still higher than many previous cohorts, but for an explanation as to why this is the case we need to bring mortgage principal (the capital repayment) into the picture.

Today’s young owners hold more mortgage debt than ever before

Mortgage principal payments are often viewed as the purchasing of an asset rather than an ongoing cost of housing and therefore we have excluded this from our analysis thus far. There are however, good reasons to bring it back into consideration. From a lived experience point of view there is little difference between the principal and interest repayments that are made on a mortgage. Both reduce the amount of income people have available for consumption, and many mortgagors may never ‘draw down’ on the asset (the value of which could, of course, fall as well as rise over time).

So what does the intergenerational picture look like if we measure the principal that owners pay as a proportion of their income? In contrast to the complex mortgage interest costs story over time, the intergenerational pattern for principal repayments is simple and striking. Figure 28 shows how each subsequent cohort of mortgagors has had to spend a greater proportion of its income on principal repayments than those before it.

intergencommission.org @resfoundation Home Affront Section 3 39

Contrast for example the last boomer cohort (those born between 1961-1965) who spent 3 per cent of their income on repayments at the age of 35, with the latest generation X cohort (born 1976-1980) who spent 13 per cent.

Figure 28: Proportion of income spent on mortgage principal by five year birth cohorts: GB Technical chart info (esp y axis) 14%

12%

10%

1926-1930 1931-1935 1936-1940 1941-1945 2% 1946-1950 1951-1955 1956-1960 1961-1965 1966-1970 1971-1975 1976-1980 1981-1985 1986-1990 0% 20 25 30 35 40 45 50 55 60 65 70

Source: RF analysis of IFS HBAI (FES) 1961 to 1991; DWP HBAI (FRS) 1994-95 to 2015-16

Note: See Figure 2

If we then bring principal into our HCIR picture over time the recent history of mortgagor HCIRs looks quite different. Figure 29 shows that, on this account, today’s young mortgagors are not faring as well as was implied by Figure 27.

intergencommission.org @resfoundation Home Affront Section 3 40

Figure 29: Proportion of income on housing costs including principal by five year birth cohorts, mortgaged owners only: GB Technical chart info (esp y axis) 30%

25%

20%

15% 1926-1930 1931-1935 1936-1940 1941-1945 1946-1950 1951-1955 1956-1960 1961-1965 10% 1966-1970 1971-1975 1976-1980 1981-1985 1986-1990 5%

0% 20 25 30 35 40 45 50 55 60 65 70

Note: All specification same as Figure 2 except housing costs are gross of principal

Source: RF analysis of IFS HBAI (FES) 1961 to 1991; DWP HBAI (FRS) 1994-95 to 2015-16

The up-front cost of buying a home has shut people out of purchasing

Even when we factor the principal payment into the HCIR measure, however, we are still presented with a conundrum: why do the home ownership rates of today’s young people lag those of previous generations when their ongoing costs of ownership are relatively low?

There is, of course, another way that affordability can affect potential home owners which is as a barrier to entry. With the exception of cash buyers and those who purchased their home in the era of 100 per cent mortgages, all first time buyers need to raise a deposit before they can join the ranks of home owners.

The size of the average deposit required over time is the product of the prevailing house prices and credit conditions. Young people keen to buy in the 1960s and 1970s benefited from low house price to income ratios, but faced limited access to credit markets. After credit was liberalised from the 1980s onwards, it was easier to get a mortgage which covered a greater part of the purchase price. But studies have shown that easy credit has been a key factor pushing up house prices over time, resulting in potential home owners chasing a moving target when it came to how much they must save before they could

intergencommission.org @resfoundation Home Affront Section 3 41

buy a home.20 The picture has changed somewhat since 2008: house prices fell briefly and are growing now at a slower rate than in the 2000s. Lending has also become more restricted, first as a result of the credit crunch and then more formally – since 2014 – following the mortgage market review (MMR).

Figure 30: Estimated number of years required to save for a first time buyer deposit among young (27-30 year old) households: UK

Technical chart info (esp y axis)

19 20

5 3

0 1983 1986 1989 1992 1995 1998 2001 2004 2007 2010 2013 2016

Note: Calculated by applying median first time buyer loan-to-value to average first time buyer house price in each year. Level of young person household savings based on putting aside 5% disposable income a year at five-year average interest rate. Appropriate stamp duty charges are added to the cost of the required deposit

Source: RF analysis of UK Finance data

Figure 30 illustrates how prices and credit have conspired to determine how long it would take a typical family headed by a 30 year old to save for an average sized deposit. For much of the 1980s it took such a family around three years; today it would take 19 years to accumulate enough to enter the market. Couple this with the fact that today’s renters must spend so much more of their income on housing costs, it is small wonder that many studies point to the rise of the ‘bank of mum and dad’ as more young people rely on family assistance to help them bridge this gap.21

20 See, for example, Oxford Economics, Forecasting UK house prices and home ownership: A report for the Redfern Review into the decline of home ownership, November 2016 21 Department of Communities and Local Government, English Housing Survey: First time buyers report 2015- 16, DCLG 2017

intergencommission.org @resfoundation Home Affront Section 3 42

In previous work for our Intergenerational Commission ‘increases in house prices’ was the number one reason why people of all ages thought today’s young adults would have a worse life than their parents.22 This section has highlighted the role of other deter- minants of housing affordability. Social rent setting, housing benefit provision, the regulation of the private rented sector and credit conditions all have a key bearing on the different housing costs we observe across the generations and the way that housing affordability has deteriorated over time. But has this also happened because what we get for our money is of better quality today? In the next Section we take on this issue.

22 H Shrimpton, G Skinner and S Hall, Millennial bug: public attitudes on the living standards of different generations, Resolution Foundation 2017

intergencommission.org @resfoundation Home Affront Section 4 43

Section 4

Housing quality over time In the previous Section we saw that generation on generation we are spending a larger proportion of income on housing in the UK. But are we getting more for our money? In this section we look at how the quality of housing has changed over time, consider what the compositional shift we noted in Section 2 means for our collective housing experience overall, and look at some of the compromises that today’s younger generations appear to be making in their quest for a home.

We should begin, however, by acknowledging the obvious: that the quality of the UK’s housing stock has improved greatly over the generations. Figure 31 presents some of the leading indicators of housing quality in England captured by the first National House Condition Survey in 1967. As this shows, more than one in five properties at that time did not have a sufficient supply of hot water, just shy of the same number did not have an inside toilet and more than one in ten were deemed unfit and either scheduled for significant renovation or demolition. On all counts, those living in the private rented sector experienced the poorest conditions.

Figure 31: Housing quality in 1967 by tenure: England Technical chart info (esp y axis)

60% All Owner occupied 50% New Town or local authority owned Private rented and other 40%

30%

20%

10%

0% % without1 inside wc % without 2hot water % deemed3 unfit

Source: RF analysis of National House Condition Survey 1967

intergencommission.org @resfoundation Home Affront Section 4 44

Fifty years on, such privations are largely a thing of the past. That is not to say, however, that issues with housing quality have disappeared in light of current norms. Figure 32 provides a similar snapshot of conditions today, showing that more than one in ten homes have no central heating, 4 per cent suffer from damp and a significant one in five fail the decent homes standard. Once again, the private rented sector is the poorest performer when it comes to decency.

Figure 32: Housing quality by tenure 2012-15: England Technical35% chart info (esp y axis) all

30% owner occupied social rented private rented 25%

20%

15%

10%

0% % with no central heating % with damp % failing decent homes standard

Source: RF analysis of English Housing Survey 2014-15

Older owners and private renters are more likely to live in poorer conditions than their young counterparts

Changing norms and a lack of a consistent data series present us with a something of a challenge when it comes to analysing housing decency across the generations. However, it is possible to look at the age breakdown of those who live in properties that fail current decent homes standards. Figure 33 does just this, showing that while it is younger headed households who are marginally more likely to live in non-decent homes than older people in the social rented sector, this pattern is broadly reversed in the other two main tenures. Close to one quarter of properties owned by someone aged 75-plus are not considered decent, in part at least because of the practical and financial challenges that home maintenance can pose in later life.23

23 See, for example, S Adams, Off the radar: Housing disrepair and health impact in later life, Care and Repair 2016 intergencommission.org @resfoundation Home Affront Section 4 45

Even more significantly, almost 45 per cent of 65-74 year olds living in the private rented sector are housed in sub-standard conditions compared to less than 30 per cent of 25-34 year olds. What explains this age gradient is an open question: do older private renters lack the bargaining power of their younger counterparts? Are they simply more reconciled to poorer conditions given the quality norms that prevailed earlier in their life course? Either way, the prevalence of low quality accommodation in the private rented sector should give us pause for thought given the prospect that more families are living in this tenure for significant parts of their lives.

Figure 33: Proportion of homes failing decent homes standard by head of household age and tenure 2012-2015: England

Technical50% chart info (esp y axis) Owner occupied 45% Social rented 40% Private rented 35%

30%

25%

20%

15%

10%

0% 19-24 years 25-34 years 35-44 years 45-54 years 55-64 years 65-74 years 75 plus

Note: No observation available for 19-24 year old owner occupiers due to small sample size

Source: RF analysis of English Housing Survey, 2010-2015

A squash and a squeeze

While the physical quality of housing stock may have improved in leaps and bounds across the generations, ongoing debates often suggest that while our homes may be a better standard we simply have less space – but is this really the case? 24

24 See, for example T Lloyd, ‘Weakening space standards won’t make homes cheaper – just smaller’, Shelter 2017

intergencommission.org @resfoundation Home Affront Section 4 46

Figure 34 looks at the proportion of households in each generation living in a property that breaches the bedroom standard (see Box 2).25 Gen Xers, baby boomers and the silent generation have all fared similarly when it comes to overcrowding at the same points in their life course. In contrast, households headed by millennials have experienced higher levels of overcrowding in their 20s than the previous generation: 9 per cent of households headed by someone approaching thirty years old today is in breach of the bedroom standard.

Figure 34: Proportion of households living in overcrowded conditions by generation (3 year rolling averages): UK Technical chart info (esp y axis)

12% Millennials (1981-2000) Generation X (1966-1980) 10% Baby boomers (1946-1965) Silent generation (1926-1945) 8%

0% 20 25 30 35 40 45 50 55 60 65 70 75

Source: RF analysis of Family Resources Survey, 1994-2016

25 In this section of the report we use the household rather than family as our unit of analysis

intergencommission.org @resfoundation Home Affront Section 4 47

i Box 2: The bedroom standard

The bedroom standard has been used as a measure of especially by those with children over the age of 10. overcrowding since the 1960s and continues to be an Conversely, those who are most likely to be deemed important reference point for policy today. The standard under-occupying are older people living in homes that is premised on a specific set of norms: that a married or once, but no longer, housed children. The standard is not cohabitating couple or any single adult aged 21 or over without its contemporary challengers however. In Shelter’s should have their own bedroom; that two siblings of 2016 Living Home Standards exercise the question of the same sex aged 10-20 could be expected to share a whether children of any gender should share bedrooms bedroom; and that it is also appropriate for two siblings was hotly debated.1 Likewise, in the Minimum Income of different sexes under the age of 10 to share. Any other Standard project participants have agreed that a spare person in the household aged 10-20 should be paired, bedroom is necessary for older couples to host visitors if possible, with a child under 10 of the same sex, or, and family as well as to sleep separately from a partner if if that is not possible, given a separate bedroom. An in poor health.2 unpaired child under 10 is also expected to have their own bedroom. 1 Shelter, Living Homes Standard, Shelter 2016 By virtue of its construction the bedroom standard is 2 A Davis et all, A Minimum Income Standard for the UK in most likely to be breached by families with children and 2016, Joseph Rowntree Foundation 2016

Looking at the usable floor space per household member over time provides a different perspective on the question. Figure 35 shows the change in space available to various household types between 1996 and 2013-15. To begin, it is worth noting that the mean amount of space overall has not changed significantly over time – or put differently, everyone’s potential share of the total housing space available has remained broadly the same. However, what is striking is the distribution of space between tenures, with each household member living in the private rented sector having on average 8m2 less today than they did in 1996. In contrast, those who live in an owned property enjoy an extra 4m2 each.

intergencommission.org @resfoundation Home Affront Section 4 48

Figure 35: Average visable floor space (m2) per household member between 1996 and 2013-15 by tenure and age: England Technical chart info (esp y axis) 1996 2013-15

All Under 45s 45+ Owners Under 45s 45+ Private renters Under 45s 45+ Social renters Under 45s 45+

20 25 30 35 40 45 50 55 60

Source: RF analysis of Survey of English Housing and English Housing Survey, 1996-2015

Given the underlying generational shifts in tenure, today’s households headed by under-45s have been net losers in the space stakes over time – each member in such a household now enjoys 3 per cent less space than they did a generation ago. In contrast, households headed by over-45s are net gainers, with members in these homes enjoying an average 4 per cent more space than they did two decades ago.

Under-occupation over the life course

Given all this, it is unsurprising that a head of steam has built up around the question of whether it is fair from an inter-generational perspective for older people to live in properties that are larger than their needs suggest. Isn’t it time to do something about this over-consumption?

Figure 36 flips the analysis and shows the proportion of households who are under-occupying according to the bedroom standard. This highlights a couple of interesting points. First, it is entirely normal not just for older generations but also younger people to have more space than their minimum needs – indeed more than half of households headed by millennials are under-occupying by their late 20s. Second, each generation is tracking the experience of its predecessor closely when on this metric suggesting that under-occupation is far more a product of the natural life course than of inter-generational change.

intergencommission.org @resfoundation Home Affront Section 4 49

Figure 36: Proportion of households under-occupying by generation (3 year rolling averages): UK Technical chart info (esp y axis)

100%

90%

80%

70%

60% Silent generation (1926-1945) Baby boomers (1946-1965) 50% Generation X (1966-1980)

40% Millennials (1981-2000)

30% 20 25 30 35 40 45 50 55 60 65 70 75 80

Source: RF analysis of Family Resources Survey, 1994-2016

While aggressive action to redistribute space from older to younger families would be both unethical and politically unpalatable, there have been calls in recent years to increase the incentives for older people to move by, for example, exempting downsizers from stamp duty. While qualitative work suggests that transaction costs are not a key reason why older people fail to downsize in significant numbers26 - and it may be questionable from an inter-generational fairness point of view to exempt those with significant wealth from this transaction tax – creative ways of facilitating downsizing are clearly an interesting area for policy to consider.27

Commuting woes

Space is not the only quality compromise that younger generations appear to be making when it comes to their homes. Figure 37 presents the mean commuting times of each generation at various age points and shows that millennials appear to be an unlucky generation on this metric too. What may look like only small increases in the daily commute add up over time: if the differences between the average commuting times of each generation were to endure, we estimate that millennials will spend 64 hours (or almost three full days) more commuting in the year they turn 40 than the baby boomers did at the same age.

26 See, for example, J Panel, H Aldridge and P Kenway. Older people’s housing; Choice, quality of life and under-occupation, Joseph Rowntree Foundation 2012 27 C D’Arcy and L Gardiner, The generation of wealth: Asset accumulation across and within cohorts, Resolution Foun- dation 2017 intergencommission.org @resfoundation Home Affront Section 4 50

Figure 37: Mean travel to work time in minutes by generations: UK Technical chart info (esp y axis)

34 Silent generation (1926-1945) Baby boomer (1946-1965) 32 Gen X (1966-80) Millennials (1981-2000) 30

20 20 24 28 32 36 40 44 48 52 56 60

Source: RF analysis of Labour Force Survey 1992 -2016

We can break this picture down by tenure to explore who within the younger generations is experiencing these longer commutes. Proximity to work is something we often think those who aspire to own a home in these straitened times may consider trading off but surprisingly we do not observe significant differences between the average travel to work times of millennial owners and private renters (Figure 39). However, millennial homeowners do have longer commutes than gen Xers at the same age – who in turn have had significantly longer commutes than the baby boomers.

intergencommission.org @resfoundation Home Affront Section 4 51

Figure 38: Mean travel to work time in minutes by generations and tenure: UK (solid lines show owners, dotted lines show private renters)

Technical chart info (esp y axis) 34 Baby boomer (1946-1965) Generation X (1966-1980) 32 Millennials (1981-2000) 30

20 20 24 28 32 36 40 44 48 52 56 60

Source: RF analysis of Labour Force Survey 1992 -2016

Finally, it is interesting to consider whether all types of earners have experienced an increase in their commute times across the generations. Figure 39 plots the time it takes to travel to work for those with in the lowest and highest earnings quartiles. While higher earners live further away from work than their lower earning peers, better off millennials appear to be tracking the experience of their parent’s generation quite closely when it comes to commute times. In contrast, lower earning young people today are spending more time commuting than gen Xers at the same age.

intergencommission.org @resfoundation Home Affront Section 4 52

Figure 39: Mean travel to work time in minutes for earnings quartile 1 (lowest) and quartile 4 (highest) by generations: UK (dotted line shows q4, solid q1)

Technical chart info (esp y axis) 45 Baby boomer (1946-1965) Generation X (1966-1980) 40 Millennials (1981-2000)

15 20 24 28 32 36 40 44 48 52 56 60

Source: RF analysis of Labour Force Survey 1992 -2016

The quality of housing is quite strained

As this section has shown, the inter-generational story with respect to housing quality is not straightforward. All generations may have benefited from the dramatic overall improvements in housing stock over the last fifty years (although older owners and private renters today are more likely to live in non-decent properties than people of a younger age). But in many other ways it is younger people who get short shrift when it comes to housing quality. Whether we look at over-crowding, space per household member or travel to work times, younger people are doing worse than their counterparts in previous generations. Once again, it is tenure that frequently intersects with age to produce the poorest outcomes for younger cohorts – and so it is to this topic that we return.

intergencommission.org @resfoundation Home Affront Section 5 53

Section 5

Looking to the future

Throughout this report the compositional shifts in tenure that we observed in Section 2, and in particular the significant rise in the share of young people living in the private rented sector, often explain the poorer housing outcomes that many have today when compared with previous generations. But what is the future likely to hold? While we recognise that home ownership isn’t for everyone and that there is no reason to suppose that any given level is ‘right’, we do know that owning a home is the preferred tenure and that it brings some definite benefits. Given this, in this final section we draw on our understanding of the conditions that have determined home ownership rates in the past produce two scenarios, and then explore the potential ramifications of each of these in turn.

We are not the first to try to undertake such an exercise. While there are no official projections of tenure trends, in recent years a number of leading academics have drawn on income, credit, price and supply forecasts to estimate future home ownership levels.28 Despite using different models and different inputs, their prognoses are often very similar: most project that between 62 per cent and 65 per cent of households will own their own homes by 2025 (according to the latest LFS data for 2017 the household home ownership rate now stands at 65 per cent).29

Playing catch up?

Our approach is slightly different. To begin, we ask whether the fall in home ownership for younger families is simply a result of the tighter conditions we observed after the financial crisis. If so, would it not be reasonable to assume that once we return to ‘normal’ this suppressed demand will be unleashed and homeownership will rise?

There are some studies which suggest that cohorts who experience low home ownership rates at a young age (say, 30) simply ‘catch up’ over time (for example, by the time they reach 40).30 If this was the case we would expect to see the home ownership rates of each birth cohort stabilise at a similar level over time. Figure 40, however, suggests we should be cautious in our hope that this ‘catch up’ effect could save the day. While there are some periods where a low starting point at 30 is followed by significant home ownership growth up to the age of 40 (for example, the cohort born in 1931-1940) this does not hold true across the board. In fact there have been periods when the starting home ownership rate at 30 and the growth rate between 30 and 40 have both been very robust (for example, the cohort born 1941-1950).

28 See for example, Oxford Economics, Forecasting UK house prices and home ownership: A report for the Redfern Review into the decline of home ownership, November 2016; Price Waterhouse Cooper, UK housing market outlook: The continuing rise of generation rent, July 2015; Cambridge Centre for Housing Policy, Housing in Transition: Understanding the dynamics of tenure change, 2012 29 There are, however, some divergent views. See for example D Miles, Could housing become like jets? Too expensive to own, Imperial College Business School, February 2017, who argues that home ownership is in secular decline. 30 See, for example, R Bottazzi, T Crossley and M Wakefield, Late starters or excluded generations? A cohort analysis of catch up in home ownership in England, IFS Working Paper W12/10, 2012

intergencommission.org @resfoundation Home Affront Section 5 54

Figure 40: Home ownership at age 30 and ‘catch up’ by age 40, by birth cohort: UK Technical chart info (esp y axis)

75%

70% Home ownership at 40 65%

60%

55% Home ownership at 30 50%

45%

40%

35%

30% 1931-40 1936-45 1941-50 1946-55 1951-60 1956-65 1961-70 1966-75

Note: Chart shows rolling ten year averages for each birth year in order to smooth trends

Source: RF analysis of Family Expenditure Survey 1961-1983; Labour Force Survey 1984-2017

Conversely, and more worryingly, if we look at more recent birth cohorts we see both their ownership at age 30 and the rate of growth between 30 and 40 drift down over time. To date those cohorts who struggled to buy in the early and mid-2000s (a period of rapidly increasing house prices) and contended with the effects of the 2008 crisis (with more limited credit and falling average incomes) have not been able to ‘catch up’ in the way that previous studies suggest they would. But is this because there is nothing special about the age of 40? Could these cohorts simply reach the same level of ownership as their predecessors but at a later age?

The home ownership window

‘Getting on the housing ladder’ often sounds like hyperbole but looking to the past does suggest that there is a window of opportunity when it comes to home ownership that closes by a certain age.

intergencommission.org @resfoundation Home Affront Section 5 55

Figure 41 makes the point, showing how home ownership levels have evolved for different five year birth cohorts over the life course. Older cohorts – especially those who benefited from Right-to-Buy – have recorded rising home ownership rates in later life. However, for more recent cohorts growth tapers off significantly by the early 40s. Put another way, in the absence of policy that facilitates purchasing in later life, for those still not owning at that age the prospect of owning one’s home beyond this point diminishes significantly.

Figure 41: Five-year cohort home ownership rates by age: UK

Technical80% chart info (esp y axis) Prime house purchase years

70% 1946-1950 60% 1951-1955

50% 1956-1960

1961-1965 40% 1966-1970 30% 1971-1975

20% 1976-1980

1981-1985 10% 1986-1990 0% 20 25 30 35 40 45 50 55 60 65 70

Notes: Results are shown only where every member of a cohort has reached that age

Source: RF analysis of Family Expenditure Survey 1961-1983; Labour Force Survey 1984-2017

Of course, there are good reasons why this may not hold true in the future and today’s young people could see their levels of home ownership continue to increase at reasonable rates well into their 40s. The rising state pension age means that many will have longer working lives than their parents and grandparents, for example, potentially extending the window of opportunity to buy a home. However, this effect could be countered by the fact that mortgages are increasingly offered on longer terms: in 2017, for example, the average first time buyer took out a mortgage with a length of 29 years compared to a 26 year term in 2005.31

Critically, there is also an open question about what will happen to the extensive housing wealth that older generations currently own as they age and die. Will this be consumed

31 Data extracted by UK Finance, September 2017

intergencommission.org @resfoundation Home Affront Section 5 56

during later life or absorbed by social care costs? Or could it be bequeathed and thereby resolve the housing issue for many of their families? Previous studies have suggested we should not over-rely on inheritance as a solution to the home ownership problem – if nothing else because it entrenches already existing inequalities – but this is a critical topic we will return to in future work for the Intergenerational Commission.32

If not now, then when?

For now, we proceed on the basis that today’s younger cohorts will not buck previous trends and will instead continue to find that age acts as a constraint on their ability to become home owners from their early-40s onwards. Using data from the past 44 years we have constructed a model that explains how prices, incomes, credit and supply have interacted at different points in time to produce the various ownership outcomes we observe for previous birth cohorts.33 While projections are never without risk, we can use our model to produce upper and lower estimates of the future home ownership rates of today’s young people.

Figure 42 presents the ‘optimistic’ picture. Here, we assume that our younger cohorts will experience the same underlying conditions over the next ten years that their counterparts witnessed in the decade with the strongest growth rate in home ownership (1981- 1991). This was a period characterised by credit liberalisation, for example, and while house prices grew faster than incomes over the decade they did so at a far slower rate than we saw, for example, in the 2000s. The ‘Right-to-Buy’ policy was also introduced in 1980 and over the decade between 1981 and 1991 around 100,000 properties per annum were sold through the policy.

32 See, for example, E Karagiannaki, Recent trends in the size and the distribution of inherited wealth in the UK , CASE 146, London School of Economics 2011 33 See Annex 3 for further details on the modelling and assumptions in this section

intergencommission.org @resfoundation Home Affront Section 5 57

Figure 42: Home ownership by cohort birth year (optimistic scenario): UK Technical chart info (esp y axis) 80%

70%

60%

50%

40% 1951 1961 30% 1971 1981 20% 1986 1991 10%

0% 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45

Source: RF modelling based on ONS, LFS, FES, and other data sources. Full details can be found in Annex 3

As the chart shows, if such conditions were repeated we could expect our millennial cohorts to reach similar levels of home ownership as generation Xers born in 1971, but to remain some percentage points shy of the levels that the baby boomers were able to reach by the age of 45. There, are however, some reasons why we think this scenario may be too optimistic. First, we are unlikely to witness a return to the credit conditions of the 1980s and 1990s – at present the average loan-to-value ratio is 84 per cent whereas between 1981 and 1991 it was 95 per cent. Second, we are very unlikely to see the same level of Right-to-Buy sales as we saw in the 1980s. Our scenario is based on the number of Right-to-Buy sales increasing by around 50 per cent, to around 30,000 a year; the average for the past decade was around 15,000.

Conversely, what would happen if the conditions we observed during the decade with the slowest home ownership growth rate (2001-2011) were to prevail? During this period the house price to income ratio rose significantly, the average loan-to-value ratio fell by 11 percentage points and Right-to-Buy sales declined by an average of 15 per cent per annum, falling to 8,100 in 2011.

Figure 43 presents this more ‘pessimistic’ scenario. Here, the inter-generational disparities in home ownership we observed in Section 2 look set to endure, with the millennial cohorts projected not just to undershoot the baby boomers but generation X as well. On this account, today’s younger cohorts are heading towards a new equilibrium home ownership rate some 10 percentage points lower than generation Xers, and a

intergencommission.org @resfoundation Home Affront Section 5 58

staggering 20 percentage points lower than that of the baby boomer generation. Again, however, there may be some reason to temper our pessimism: the government’s current house building targets, for example, are higher than the assumptions we model here.

Figure 43: Home ownership by cohort birth year (pessimistic scenario): UK Technical chart info (esp y axis)

80%

70%

60%

50%

40% 1951 1961 30% 1971 1981 20% 1986 10% 1991

0% 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45

Source: RF modelling based on ONS, LFS, FES, and other data sources. Full details can be found in Annex 3

The impact of deferred home ownership

While reality is most likely to lie somewhere between our upper and lower bounds, we should not be sanguine about even the ‘optimistic’ scenario we present. While in this scenario millennials may approach similar levels of home ownership to previous generations by the age of 45, the fact that they are reaching this at a later point in their life course has two key implications.

To begin, later mortgaged home ownership is likely to mean later outright home ownership, a trend that could be further exaggerated by mortgages more commonly being extended on longer terms. Unlike their older peers, today’s young people could find themselves paying off a mortgage well into their 50s and 60s, and possibly beyond

intergencommission.org @resfoundation Home Affront Section 5 59

working age. Not only does this mean less time as an owner to benefit potentially from rising house prices; this could also have a serious impact on the ability of such cohorts to save for retirement during what have been, to date, prime savings years.

Second, later home ownership could mean more children spending their early years in other tenure types than in previous generations. The average age at which women are having their first child has, of course, risen over time – those born in 1961 typically had their first child at the age of 24 while today the average age is 29. Yet even when we factor in later parenthood and use our ‘optimistic’ scenario, Figure 44 shows that fewer millennials will own their home by the time their first child is five than previous generations, with all the implications for security and quality that this implies.

Figure 44: Home ownership by birth cohort and age (optimistic scenario) and parenthood: UK Technical chart info (esp y axis) 80% Age at which Age at which average baby average 70% boomer's first millennial's first child turns 5 child turns 5 60%

50%

40% 1951 1961 30% 1971 1981 20% 1986 1991 10%

0% 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45

Note: Age on chart refer to head of family; parenting age refer only to women

Source: RF modelling based on ONS, LFS, FES, and other data sources. Full details can be found in Annex 3

intergencommission.org @resfoundation Home Affront Section 5 60

Renting as the new normal?

Our pessimistic scenario, of course, gives even more pause for thought. If these projections were to hold true, millennials will not reach ownership rates of 50 per cent-plus before the age of 45 (Figure 45) and renting a home could become the new normal. But would this matter? After all, there are plenty other parts of the world (and indeed periods of time) where home ownership is not the dominant tenure.34

Figure 45: Home ownership by birth cohorts and the majority (pessimistic scenario) and the majority: UK Technical chart info (esp y axis)

80%

70%

60% Majority of families are home owners 50%

40% 1951 1961 30% 1971 1981 20% 1986 1991 10%

0% 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45

Source: RF modelling based on ONS, LFS, FES, and other data sources. Full details can be found in Annex 3

There are a number of reasons why we should be concerned by a lower home ownership equilibrium than that enjoyed by previous generations.

• First, home ownership is much the preferred tenure in the UK and families’ expec- tations of buying a home remain consistently high.35 As far as choice goes, renting (at least in the private rented sector) is clearly seen as sub-optimal. While norms can shift over time, in the immediate future disappointment for a large part of the population will pose a significant political challenge.

34 See, for example, A Alik-Lagrange and T Schmidt, The pattern of home ownership across cohorts and its impact on the net wealth distribution: empirical evidence from Germany and the US, Deutsche Bundesbank 2015 for an interesting take on the comparative home ownership experience by age cohorts 35 Department for Communities and Local Government, English Housing Survey: Future home owners, 2015-16, DCLG 2017

intergencommission.org @resfoundation Home Affront Section 5 61

• Second, tenure can have a huge bearing on one’s cumulative housing costs across the life-course. With its high housing cost to income ratios (HCIR) the private rented sector is the least affordable tenure in which to live: today’s 30 year olds who rent privately spend 30 per cent of their income on housing (net of housing benefit) while their owning peers spend just half of this (15 per cent). Rolled up over a lifetime, these disparities will constitute a significant drag on living standards.

• Third, as previous work for the Intergenerational Commission has shown, home ownership has been the primary way that previous generations have built up wealth over the course of their lifetime, both as result of the enforced saving it constitutes and the windfall gains it has delivered as house prices have risen.36 Today’s young people will need to find new savings mechanisms that substitute for home ownership if their future levels of wealth are to be guaranteed.

On any account, the future looks very different for young people when it comes to housing

Regardless of what the future brings, both of our scenarios suggest that today’s young people face a very different housing outlook to that of previous generations. Whether home ownership is temporarily deferred or simply no longer on the cards, life in the private rented sector looms far larger for millennials than it has for their parents with huge implications for their sense of security and their quality of life.

That is not to say there is no scope for change however. While some underlying conditions (for example, the cost of credit) are not within government’s control, many other of the factors that determine a generation’s housing outcomes are within the gift of politicians. Just how policy makers could respond to the significant challenge presented by the ‘housing affront’ is a topic to which we will return in future Intergen- erational Commission analysis.

36 C D’Arcy and L Gardiner, The generation of wealth: Asset accumulation across and within cohorts, Resolution Foun- dation 2017

intergencommission.org @resfoundation Home Affront Section 6 62

Section 6

Conclusion

Housing is a source of great anxiety for many today – both on their own account and in their concern for others. Old and young, higher income and lower income, the quest for a secure, affordable and decent home is something that exercises the many. While there have been good news stories over the last fifty years – most notably the significant improvements in the quality of housing stock – deteriorating housing affordability and significant shifts in tenure make housing that rare thing: a true majoritarian problem.

Throughout this report, however, we have shown that it is today’s younger people who are truly at the sharp end of the housing crisis. Despite spending a larger share of their income on housing costs than previous generations did at the same age, millennials get far less in return. They are far more likely to rent privately than young people did in the past, with all the insecurities attached to that tenure type. Moreover, they have less space and live further from their workplaces than previous generations at the same age.

This changing picture has undeniable consequences in the here and now – higher housing costs leaves young people with less disposable income and less opportunity to save. But it also has significant consequences for today’s young people over the whole of their life course – will they have to bring up their children in insecure homes for example, or continue paying off a mortgage into retirement? Will the housing outcomes they achieve become more contingent on their parents’ wealth than ever before?

That is not to be without hope however. There is a strong political imperative to address the housing problem. In the final months of the Intergenerational Commission, we will return again to the question of housing and consider what action policy makers could take to address the many issues we have documented here, as well as the tricky question of inheritance.

intergencommission.org @resfoundation Home Affront Annex 1 63

Annex 1

Our housing tenure categories and data

There are many possible categorisations of tenure. Traditionally, and in the government’s English Housing Survey (EHS) – which, as the name suggests, only covers England – the unit of analysis is the dwelling. By the usual definition, a property may be occupied by:

1. Owner occupiers who own outright (regardless of any lodgers);

2. Owner occupiers who are buying with a mortgage (including those with part-rent arrangements);

3. Private renters (including those living rent-free);

4. Social renters renting from the local authority; or

5. Social renters renting from the housing association.37

Note that most surveys, including those used in this report, generally exclude those living in institutional residences such as care homes, prisons, student halls and army barracks. These are not insignificant omissions. Census figures show that over 400,000 people resided in care homes in Great Britain in 2011,38 there are around 90,000 people in prison in the UK, and in term-time there are around 500,000 students in halls.39

By this conventional measure we can calculate that there are around 27 million occupied dwellings in the UK. However, by counting dwellings without looking at the number of people in each, their relationship to each other or their housing aspirations, this approach risks missing a large number of people and significant changes and pressures in housing.

As we’ve said previously, “Imagine, for example, a person who buys a house but who then takes in three lodgers. On the standard measure, this simply counts as an owner-occupied household – and the three residents that rent drop out of the picture. Or consider an adult returning to the parental home. That individual also disappears from the statistics. And five unrelated people who share a house? They would be counted as one rented household rather than the five separate renters that most would intuitively regard them to be.”40

37 The EHS notes that “a large number of HA tenants wrongly report that that they are LA tenants; most commonly because their home used to be owned by the council but had transferred to a housing association.” 38 ONS, Labour Force Survey User Guide, Volume 1 – LFS background and methodology 2015 39 https://www.hesa.ac.uk/data-and-analysis/students 40 L Judge & A Corlett, Only half of families own their own home – how do the other half live?, Resolution Foundation blog, December 2016

intergencommission.org @resfoundation Home Affront Annex 1 64

To improve on this, we have a developed a method that distinguishes between different ‘family units’ – each single adult or couple, plus any dependent children – in the same dwelling. This might also be considered an assessment of the number of potential homebuyers and current owners. We divide families into nine tenure categories:

1) Owner occupiers who own outright;

2) Owner occupiers who are buying with a mortgage (including those with part-rent arrangements);

3) Social renters renting from the local authority;41

4) Social renters renting from a housing association;42

5) Private renters (including those living rent-free) who have a household to themselves;

6) Private renters who share a household with other private renters (which can include relatives);

7) Single childless adults living in their parents’ home (age 19+) and not studying full-time;43

8) Single childless adults living in their parents’ home (age 19+) but studying full-time (excluding those in student halls);44 and

9) Other family units in someone else’s owned or social rented home. This includes non-single, or non-childless adults living in their parents’ home with their partner/children; elderly people living in their children’s home; other relatives; and unrelated lodgers (whether they pay rent or not).45

Where an adult (age 19+) is living in their parents’ home, for example, this would be counted as two family units and a home ownership rate of 50 per cent – rather than the single 100 per cent owner-occupied home that dwelling-based analysis would say.

Some of these choices and divisions are subjective and some are driven by what data is available. But, while other choices or finer discrimination would be possible – including counting the total population or total adult population, rather than number of families – the above allows us to paint a rich picture of what’s happened to tenure without descending into a confusing array of small categories.

41 Including new town development corporations in the 1980s. 42 In the data we use, housing associations are not coded separately prior to 1980 and are instead included in the PRS. 43 We exclude those under 19, on the basis that many 18 year olds who would otherwise be included would in fact still be in compulsory education or have only recently finished. Parents can include step-parents. 44 Prior to 1984, we are unable to split out full-time students from other adults living in their parents’ home. 45 Squatters are excluded from our analysis but are very few in number. Those who are entirely homeless are also excluded.

intergencommission.org @resfoundation Home Affront Annex 1 65

Data sources

Our primary data source for housing tenure is the Labour Force Survey (LFS). Although most commonly used for employment statistics, as a large, detailed, regular, timely household survey it is also a great resource for housing tenure. As we are looking at families, we use the ‘household’ rather than ‘adult’ datasets and weights.

Prior to 1984, we use the Family Expenditure Survey (FES), as standardised by the Institute for Fiscal Studies.46 There are therefore discontinuities in our data source and methodologies between 1983 and 1984. However, as Figure 46 shows, there is a strong overlap of results between the two sources at this point. The sample size available is also considerably smaller in FES, i.e. in our pre-1984 data, and especially so pre-1968, leading to noisier results.

Where we have looked at tenure by household income, the Department for Work and Pensions’ Households Below Average Income (HBAI) dataset has been used, in addition to FES. Again there is a strong agreement between sources.

Figure 46: Proportion of families (singles or couples) owning their own home in three different data sources Technical chart info (esp y axis) FES (1961-1989, UK) LFS (1984-2017, UK) HBAI (1994-2015, GB) 60%

55%

50%

45%

40%

35%

30%

25% 1961 1964 1967 1970 1973 1976 1979 1982 1985 1988 1991 1994 1997 2000 2003 2006 2009 2012 2015

46 A Goodman & S Webb, Institute for Fiscal Studies Households Below Average Income Dataset 1961-1991, 1995

intergencommission.org @resfoundation Home Affront Annex 2 66

Annex 2

Measuring affordability

Our key measure of housing affordability is the Housing Cost to Income Ratio (HCIR). This is calculated as housing costs divided by disposable income: the share of a household’s disposable income that is spent on housing. We then report the average HCIR amongst all the family units in the relevant population.

Note that to avoid outliers skewing the results we cap individual HCIRs at 100 per cent, though a small minority in the data appear to spend more than their entire income on housing costs.

As discussed in Annex 1, our sources of income data – and housing costs – are the Family Expenditure Survey (FES) and the Households Below Average Income (HBAI) dataset.

Housing costs in these datasets are: • mortgage interest;

• rent;

• structural insurance premiums;

• water and sewerage costs; and

• ground rent or service charges.

Council tax payments are not included as a housing cost.

As explored in Section 3, and in a previous report – The housing headwind47 – there are alternatives to the definition above. Firstly, we may include mortgage principal (capital) repayments within housing costs, rather than considering only interest costs. These repayments may in theory be seen as a form of saving rather than a cost, but from the perspective of people’s day-to-day living standards they may feel just as real as any other housing cost. As explored in the text, given large increases in house prices, whether or not they are included makes a material difference to our assessment of mortgagor housing pressures. Note that we are not able to assess deposits – or the costs of saving for them – as a housing cost.

Secondly, there is a question over how housing benefit should be treated. Our standard approach is to follow the definitions used in the available datasets, which treat housing benefit as income and the rent that this pays for as a cost to the household. However, given that households may not ever see this cash, it can also be illustrative to consider their incomes excluding housing benefit while considering only the housing costs that they themselves must pay. If we remove housing benefit from the calculations in this way then we find, for example, that social renters spend a lower proportion of their income on housing than in the standard approach.

47 S Clarke, A Corlett and L Judge, The housing headwind: The impact of rising housing costs on UK living standards, Resolution Foundation, 2016 intergencommission.org @resfoundation Home Affront Annex 3 67

Annex 3

Modelling of homeownership rates

To produce some scenarios for the homeownership rates of different cohorts ten years into the future we develop the approach used by Renata Botazzi, Thomas Crossley and Matthew Wakefield in their 2012 paper.48 In this paper the authors use household surveys (the Family Expenditure Survey/Expenditure and Food Survey – FES/EFS) to create a synthetic panel of cohorts and homeownership rates over time. We do the same using FES and the Labour Force Survey (LFS). This provides us with a dataset that contains the homeownership rates at ages 20 through to 80 of cohorts born between 1900 and 1991.49 We do this for cohorts who own their home in London and those that own outside the capital to account for the starkly different house price and ownership rates in London.

To this we merge a range of macroeconomic data on house prices, Right to Buy (RTB) sales, credit conditions, home building, and household income. The variables and their sources are as follows:

• Credit conditions index (Fernandez-Corugedo and Muellbauer50)

• House prices (DCLG)

• Household income (FES and HBAI)

• Home building (DCLG)

• RTB sales (DCLG)

48 R Bottazzi, T Crossley and M Wakefield, Late starters or excluded generations? A cohort analysis of catch up in home ownership in England, IFS Working Paper W12/10, 2012 49 Obviously we do not have a complete dataset. For those born in the latest year – 1991 – we only have their homeownership rates at 25 (i.e. in 2016/2017). 50 Fernandez-Corugedo, E. and J. Muellbauer, Consumer credit conditions in the United Kingdom, Bank of England Working Paper 314, 2006

intergencommission.org @resfoundation Home Affront Annex 3 68

We model the homeownership rate for cohorts at four different ages; 30, 35, 40, and 45, using the macroeconomic data above and the homeownership rate of the same cohort five years earlier. For example to model homeownership at 30 we construct the following model:

Home ownership rate at 30 is a function of:

• That cohort’s home ownership rate at 25

• The average house price to income ratio between 26-30

• Average RTB sales between 26-30

• Average house building between 26-30

• Credit conditions index between 26-30 The home ownership rate and house price to income ratio variable are interacted with a London dummy to take into account the different price, income and ownership dynamics in the capital.

Our preferred specification is a generalised linear model where the dependent variable has been transformed to take the natural logarithm of the odds ratio of the homeownership rate. This takes into account the fact that our data is bounded at 0 and 1 (i.e. home ownership cannot be lower than zero or greater than 100 per cent). We run this model on the ownership rate at 30, 35, 40 and 45. Results are provided below.

Table 1: Marginal effects of explanatory variables for our four home ownership models Dependent variable (home ownership at age) 30 35 40 45 London dummy -0.14 -0.12 -0.09 -0.07 Home ownership rate (t-5 years) 0.02 0.03 0.17 0.15 House price to income ratio -0.0007 -0.0005 -0.0004 -0.0004 RTB sales 0.011 0.015 0.012 0.004 House building 0.38 0.23 0.16 0.06 Credit conditions index (higher = more liberal) 1.63 1.43 1.55 2.40 Observations 88 88 88 88 R2 0.89 0.89 0.92 0.82 Coefficients are expressed as marginal effects where a 1 unit change in independent variable is associated with marginal effect change in the dependent variable. Marginal effects in bold are significant. All coefficients are of the expected sign and most are significant across all models. The exception is the home ownership rate five years previous which is not significant in three of the four models. The London dummy behaves as expected, although the table above only shows the marginal effect of the dummy not the interactions with the house price to income ratio or the home ownership rate five years previous.

In order to project forward homeownership rates for recent cohorts (see Section 5), we need projections for our explanatory variables. We look at two scenarios using the

intergencommission.org @resfoundation Home Affront Annex 3 69

growth rates of our variables from two specific periods: 1981-1991 and 2001-2011. These two decades were, respectively, the decade in which ownership increased at the fastest rate for those aged 30 to 45 and the decade in which it rose at its slowest rate, and for which we have data on all our explanatory variables. The table below shows the average growth rate of our explanatory variables over the two periods.

Table 2: Average annual percentage change in key explanatory variables in the two time periods that underlie our ‘optimistic’ and ‘pessimistic’ scenarios

1981-1991 2001-2011 House price to income ratio 0.5% 1.9% RTB sales 0.4% -9.7% House building -1.0% -2.0% Credit conditions index (higher = more liberal) 9.2% -4.6%

As the table makes clear the house price to income ratio, although rising in both periods, rose considerably faster in the later period. RTB sales increased in the 1980s but fell in the 2000s. Housebuilding fell in both periods but at a greater rate in the later period, and credit conditions tightened significantly in the 2000s. The result is that the 1980s provided a much more benign environment for rising homeownership than the 2000s and these are reflected in our projections.

intergencommission.org @resfoundation Resolution Foundation is an independent research and policy organisation. Our goal is to improve the lives of people with low to middle incomes by delivering change in areas where they are currently disadvantaged. We do this by:

»» undertaking research and economic analysis to understand the challenges facing people on a low to middle income;

»» developing practical and effective policy proposals; and

»» engaging with policy makers and stakeholders to influence decision-making and bring about change.

For more information on this Report, contact:

Lindsay Judge Senior Policy and Research Analyst [email protected] 020 3713 5801

intergencommission.org @resfoundation